AI Prompt Attack Taxonomy Dashboard

Precision-classified AI jailbreak & prompt injection techniques — relevant posts only · 100 rows/page

Total Techniques
2482
With Actual Prompts
2458
High Severity
2367
Categories Covered
9
Last Updated
2026-02-23 12:54 UTC
Taxonomy
All2482
Role & Persona Manipulation2043
Instruction Hierarchy Attacks409
Indirect & Prompt Injection18
Divide & Conquer / Multi-turn4
Model Extraction3
Defense & Red-team Research2
Encoding & Obfuscation1
Fictional Framing1
Social Engineering1
Severity
High
Medium
Low
Info
Sev ↕ Category ↕ Technique Name ↕ Persona/Role ↕ Technique Description Example Prompt Source ↕ Date ↕